So it’s entirely possible that you’ve heard the latest ‘groundbreaking’ finding in the religious/atheist culture war, suggesting that religion is negatively associated with compassionate prosocial behaviour:
“Love thy neighbor” is preached from many a pulpit. But new research from the University of California, Berkeley, suggests that the highly religious are less motivated by compassion when helping a stranger than are atheists, agnostics and less religious people.
In three experiments, social scientists found that compassion consistently drove less religious people to be more generous. For highly religious people, however, compassion was largely unrelated to how generous they were, according to the findings which are published in the most recent online issue of the journal Social Psychological and Personality Science.
I don’t think anyone who’s been reading this blog for longer than, say, two weeks would have any trouble pinning down where I am on the believer/non-believer scale: I am doing pull-ups from the far side of the non-believer scale, encouraging others to jump off as I dangle over the edge. I am, in fact, an unapologetic anti-theist – I think that religious faith is an inherently harmful byproduct of bad brains and lack of intellectual curiosity. That being said, I am reflexively skeptical of any scientific finding that seems too good to be true. This is definitely one of them.
So, as I have before, I am taking my super skepty powers to the paper (frustratingly behind a paywall) to see if I can spot any errors.
Saslow and colleagues hypothesize that while both religious and non-religious people are likely to engage in prosocial behaviours (doing things for other people without the expectation of remuneration), their motivations may differ. Specifically, the authors suppose that compassion/sympathy for others is a strong motivator for non-religious folks, but not for religious ones. Put another way, atheists are good to others out of compassion, whereas theists do it for other reasons.
If the hypothesis is correct, then manipulating someone’s feelings of compassion will have one of two responses. If the person is strongly religiously affiliated, they will be no more (and no less) likely to do things for other people – their compassion for others will be meaningless. If the person has no religious identity, however, they will be more likely to help out.
In order to test their hypothesis, the authors conducted three separate evaluations. In the first study, they tested whether the level of ‘trait compassion’ (how compassionate you are generally) is associated with prosocial behaviour as a function of religious belief. They did this by running statistical tests on data from a representative survey of Americans, which has measures of compassion, religious affiliation, and a list of various prosocial activities (e.g., giving to the homeless, returning change if given too much, volunteering for charity, 7 others). The survey also had data on gender, political orientation and education, which are all potential confounders of this relationship.
A second study was conducted to investigate the possible causality of the compassion/religiosity interaction in prosocial behaviour. A crucial element of causality is establishing a temporal sequence (i.e., does cause happen before the effect?), and so two groups were asked to watch one of two videos. The ‘neutral’ video was 46 seconds of two men talking; the ‘experimental’ video was a 46-second evocative piece about child poverty with images and moving music played overtop. The groups were then asked how much money (of $10 given to them) they would be willing to donate to a stranger. Attitude toward charitable giving (i.e., how much of one’s income should one donate to charity?) was also measured.
In the third and final study, the hypothetical element of prosocial behaviour was removed by creating a real-world co-operative exercise. Participants, asked about their levels of trait compassion and their religious affiliation, were partnered up and asked to play a game in which they could divy up rewards to be spread among their partners, or keep all the points. Study points could be exchanged for money. Related exercises designed to measure one’s response to the generosity of others was also conducted.
In all three studies, the main hypothesis was supported: compassion was a motivating factor in charitable giving in non-religious people; this effect was not seen (or was less pronounced) in religious people. Trait compassion in general predicted prosocial behaviour, such that those who described themselves as generally compassionate were more likely to help others regardless of their religious affiliation. In study 2 religious affiliation was associated with generosity, such that those who reported higher levels of religiosity were more likely to help others – this effect was not seen in the other two studies.
So does this study suggest that non-religious people are better than religious ones? As much as we’d love for that to be the case, this study does not lead us to that conclusion. What it says is that religious people engage in prosocial behaviours for different reasons than non-religious ones. Appeals to compassion are far more likely to work in a non-religious population, because they (we) respond to them. Whatever the reasons for prosocial behaviour in the religious (which were not examined in this study), compassion does not appear to be one of them. People in general, regardless of their religious feelings, respond to things that trigger the compassionate response – there appears to be more going on in religious populations.
The cynical interpretation of these results is that, for all their talk of ‘mercy’ and ‘loving their neighbour’, religious people do not give out of the goodness of their hearts – they do it either because they are told to or because they’re trying to get a reward*. While it’s fun to beat up on the faithful, I simply cannot arrive at that conclusion. What is far more likely is that the repeated exhortations to give to others creates a psychological attitude toward giving that has a bunch of different variables in it. Compassion is still in there, but it’s fighting with a bunch of other things – group affiliation, moralistic instruction, duty to give – that drown out (or at least moderate) the main effect of that one motivation. It would be interesting to see a study that parsed out the other reasons for prosocial behaviour in these two populations.
These studies are all pretty interesting, but they just scratch the surface of the reasons why religious folks and non-religious ones respond to altruistic cues differently. We can’t draw much stronger conclusions from this than non-religious people are more susceptible to compassionate manipulation to behave prosocially. I personally think that’s a good thing, provided our compassion is tempered with a bit of skepticism to make sure we don’t throw money at every sob story we’re presented with.
Like this article? Follow me on Twitter!
*It is, I suppose, possible that there is a certain amount of ‘just world fallacy’ going on, wherein religious people are more likely to reflexively assume that people get what they deserve, and therefore it is not their responsibility to act compassionately. I am skeptical of this idea as well, and would need to see some evidence to accept it.